Abstract Alzheimer's disease (AD) is the most common form of dementia. It currently affects 5 million people in the US, a number that is expected to rise to a staggering 16 million by 2050. AD not only deprives patients of their basic mental functions, but severely batters families and caregivers. Its costs are currently estimated at $236 billion, and will likely increase to more than $1 trillion by 2050. As our society rapidly ages, the need for combating AD is pressing. Histological and imaging studies in AD patients and animal models have shown that the entorhinal cortex is a primary site of atrophy and activity loss in the early phases of AD. However, it is still largely unclear what type of activity is lost in the entorhinal cortex in early AD. Using in vivo neurophysiological recording methods, we recently demonstrated that gamma oscillations, a network activity reflecting summed neuronal membrane potentials, are impaired in the entorhinal cortex of an AD mouse model. Our results and recent literature suggest a possibility that entorhinal gamma oscillations can be used for both a biomarker and a therapeutic target of AD. Here we propose studies to investigate the role of gamma oscillations of entorhinal cortex in memory impairments using AD mouse models. Our approach involves in vivo recording of local field potentials (theta and gamma oscillations) and spike activity, optogenetic and chemogenetic methods, closed- loop stimulation, cell-type specific histological analyses of neuronal loss and a novel APP knock-in mouse model. There are three Specific Aims: (Aim 1) identify the extent and time course of entorhinal cortex (EC) gamma impairments; (Aim 2) determine whether the reactivation of network activity using gamma stimulation of EC attenuates or eliminates memory impairments in APP-KI mice, and (3) determine cell types that underlie the EC gamma impairment. If successful, our studies will identify in vivo network mechanisms of memory impairment in AD, and will help identify neuronal activities as therapeutic targets to prevent or slow the progression of disease. Furthermore, our study will help us develop more effective and safer procedures for deep-brain stimulation as a powerful tool to preserve or improve memory function that may eventually be used to slow the rate of memory decline in AD patients.